Fiscal Policy, Inequality and the Poor in the Developing World
Nora Lustig Tulane University
Nonresident Fellow CGD and IAD
National Treasury and World Bank Workshop Fiscal Policy and Redistribution in an Unequal Society
Pretoria, South Africa, November 5, 2014
When using material in this ppt please cite as:
• Lustig, Nora. 2014. “Fiscal Policy, Inequality and the Poor in the Developing World. Round 1.” CEQ Working Paper No. 23, Center for Inter-American Policy and Research and Department of Economics, Tulane University and Inter-American Dialogue, forthcoming.
2
www.commitmentoequity.org
CEQ Teams (Year of Survey; C=consumption & I=income)(MWB Version)
1. Argentina (2009, I): Nora Lustig and Carola Pessino (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3
2. Armenia (2011; I): Stephen Younger and Artsvi Khachatryan (May 31, 2014; paper)
3. Bolivia (2009; I): Veronica Paz Arauco, George Gray-Molina, Wilson Jimenez and Ernesto Yañez (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3
4. Brazil (2009; I): Sean Higgins and Claudiney Pereira (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3
5. Chile (2009, I): Jaime Ruiz-Tagle and Dante Contreras (Oct. 25, 2014)
6. Colombia (2010, I): Marcela Melendez, Nora Lustig and Valentina Martinez (May 2014)
7. Costa Rica (2010; I): Pablo Sauma and Juan Diego Trejos (February 2014; paper)
8. El Salvador (2011; I): Margarita Beneke, Nora Lustig and Jose Andres Oliva (March 11, 2014)
9. Ethiopia (2010/11; C): Ruth Hill, EyasuTsehaye, Tassew Woldehanna (Sept. 28, 2014)
10. Guatemala (2011; I): Maynor Cabrera, Nora Lustig and Hilcias E. Moran (August 27, 2014)
10 Indonesia (2012; C) : Jon Jellema and Matthew Wai-Poi (Sept. 9, 2014)
11 Jordan (2010; C) : Morad Abdel-Halim, Shamma Adeeb Alam, Yusuf Mansur, Umar Serajuddin, Paolo Verme (May 16, 2014)
12 Mexico (2010; I): John Scott (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3
13 Peru (2009; I): Miguel Jaramillo (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3
14 South Africa (2010; I): Ingrid Woolard, Precious Zikhali, Mashekwa Maboshe, Jon Jellema (Aug. 25, 2014)
15 Sri Lanka (2009/10; C): Nisha Arunatilake, Gabriela Inchauste and Nora Lustig (April 8, 2014; paper)
16 United States (2011; I): Sean Higgins, Nora Lustig, Whitney Ruble and Timothy Smeeding (paper Oct. 2014)
17 Uruguay (2009; I): Marisa Bucheli, Nora Lustig, Maximo Rossi and Florencia Amabile (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3
3
Handbook
• Lustig, Nora and Sean Higgins. 2013. Commitment to Equity Assessment (CEQ): Estimating the Incidence of Social Spending, Subsidies and Taxes. Handbook. CEQ Working Paper No. 1, Center for Inter-American Policy and Research and Department of Economics, Tulane University and Inter-American Dialogue, September.
4
Commitment to Equity Assessments (CEQ)
• Accounting Approach: no behavioral, no general equilibrium effects and no intertemporal effects
• Point-in-time
• Mainly average incidence; a few cases with marginal incidence
• Comprehensive standard fiscal incidence analysis of current systems
• Harmonized definitions and methodological approaches to facilitate cross-country comparisons
• Uses income/consumption per capita as the welfare indicator
• Tax shifting assumptions are the standard ones
• Allocators vary => full transparency in the method used for each category, tax shifting assumptions, tax evasion
• Secondary sources are used to a minimum
• Handbook (Lustig and Higgins, 2013)
5
6
Basic elements of standard fiscal incidence
• Before taxes and transfers income of unit h, or Ih
• Taxes Ti – personal income taxes; contributions to social security – consumption and production taxes and subsidies
• Transfers Ri – social spending: cash & near-cash transfers; in-kind transfers
(education and health) – consumption and production (agriculture) subsidies
• “Allocators” of tax i and transfer j to unit h, or Sih , Sjh (the share of tax i borne or transfer j received by unit h) => Incidence
• Post-taxes and transfers income of unit h (Yh)
7
• Post-taxes and transfers income of unit h (Yh) is:
Yh = Ih - ∑i TiSih + ∑j RjSjh
8
Pre-fisc Income
Post-fisc Income
Taxes & Transfers
Incidence of Taxes & Transfers
9
MARKET INCOME
MINUS DIRECT TAXES
DISPOSABLE INCOME
NET MARKET INCOME
PLUS DIRECT TRANSFERS
GROSS INCOME
PLUS DIRECT TRANSFERS MINUS DIRECT TAXES
MINUS NET INDIRECT TAXES
POST-FISCAL INCOME
PLUS MONETIZED VALUE OF PUBLIC SERVICES: EDUCATION & HEALTH
FINAL INCOME
Construction of Income Concepts
Methods to Construct Income Concepts
• Direct Identification Method
• Imputation Method
– Direct (Education and Health)
– Simulation (Direct and Indirect Taxes)
• Inference Method
• Alternate Survey
• Secondary Sources Method
10
Reconciling the Two Economies: Survey Data vs. Administrative Accts.
• What to do when totals in Survey do not match administrative accounts?
• Should imputed values be scaled-down or the rest of the concepts scaled up?
11
Contributory Pensions
• Are they a government transfer or deferred consumption and hence part of market income?
– No consensus
– Results, especially for poverty, are extremely sensitive
=> Do it both ways
12
Importance of Comprehensive Analysis
• Obvious reason: to capture the full effect of the net fiscal system
• More subtle reason: partial assessments of progressivity and regressivity can be misleading
=> a regressive tax can be equalizing and re-inforce the equalizing impact of transfers
13
Lambert’s Conundrum
14
1 2 3 4 Total
Original income x 10 20 30 40 100
Tax Liability t(x) 6 9 12 15 42
Benefit level b(x) 21 14 7 0 42
Post-benefit income 31 34 37 40 142
Final income 25 25 25 25 100
Source: Lambert, 2001, Table 11.1, P. 278
Lambert’s Conundrum • The Reynolds-Smolensky (R-S) index for taxes in
this example is equal to -0.0517, highlighting their regressivity.
• Yet, the R-S for the net fiscal system is 0.25, higher than the R-S for benefits equal to 0.1972.
• If taxes are regressive vis-à-vis the original income but progressive with respect to the less unequally distributed post-transfers (and subsidies) income, => regressive taxes exert an equalizing effect over an above the effect of progressive transfers.
• Two renowned studies found this type of result in the US and the UK. 15
Results
Redistribution and Inequality Reduction
16
Redistribution in the rich and developing countries
Sources: EUROMOD for EU,Higgins et al. (2014) for US and for CEQ countries see Lustig (2014) and references at the end.
Note: in these calculations contributory pensions are part of market income and NOT treated as a government transfer. 17
-0.30
-0.25
-0.20
-0.15
-0.10
-0.05
0.00 Gu
ate
ma
la(2
01
0)
Co
lom
bia
(20
10
)
El S
alv
ad
or (2
011
)
Bo
livia
(20
09
)
Peru
(20
09
)
Co
sta R
ica
(20
10
)
Mex
ico
(20
10
)
Po
lan
d
Bu
lga
ria
Uru
gu
ay
(20
09
)
Bra
zil(2
00
9)
Ch
ile(2
00
9)
Greece
Lith
ua
nia
Esto
nia
Cy
pru
s
Ro
ma
nia
La
tvia
Slo
va
kia
Italy
Hu
ng
ary
Ma
lta
US
(20
11
)
Czech
Rep
ub
lic
So
uth
Afr
ica
(20
10
)
Sp
ain
Sw
ed
en
Den
ma
rk
Slo
ven
ia
Po
rtu
ga
l
Au
stria
Neth
erla
nd
s
Germ
an
y
Fra
nce
Fin
lan
d
Lu
xem
bo
urg
Belg
ium
Un
ited
Kin
gd
om
Irela
nd
ChangeinGini:Disposablevs.Market(inGINIpoints)
Redistribution in Middle and Low Income Countries: CEQ 16
18
-0.09
-0.08
-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0.00 Guate
mala(2
010)
Colombia(2
010)
ElSalvador(2
011)
Bolivia(2
009)
Peru(2009)
CostaR
ica(2010)
Mexico
(2010)
Urugu
ay(2009)
Brazil(2
009)
Chile(2009)
SouthAfrica(2
010)
ChangeinGini:Disposablevs.Market(inGINIpoints)
19
0
2000
4000
6000
8000
10000
12000
14000
16000
0.3
0.4
0.5
0.6
0.7
0.8
Sou
thA
frica(20
10
)
Brazil(2
00
9)
Co
lom
bia(2
01
0)
Ch
ile(20
09
)
Mexico
(20
10
)
ChangesinGini:DisposableVSMarketIncome
GiniofDisposableIncome GiniofMarketIncome GNIpercapita(2005PPP)
The impact of direct taxes and transfers on inequality (Gini coefficient): CEQ 16
20
0
2000
4000
6000
8000
10000
12000
14000
16000
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
SA(2010)
Bra(2009)
Col(2010)
Chi(2009)
Mex(2010)
ChangesinGiniCoef icients
GiniofMarketIncome GiniofDisposableIncome GiniofPost- iscal GNIpercapita(2005PPP)
The impact of net indirect taxes on inequality (Gini coefficient): CEQ 16
21
GTMBOL
SLV
PER
CRI
MEX
URY
BRACHL
COL
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0 2000 4000 6000 8000 10000 12000 14000 16000
GNI/Capita(2005PPP)
ChangeinGini:Post-fiscalvs.Market(declineinGinipointsshowninposi vequadrant)
Lindert’s (2006) historical result is also found in cross section: Higher GDP/capita, more redistribution
South Africa
22
However, no Robin Hood Paradox And results do not depend on South Africa
GTMBOL
SLVPERCRI
MEXURY BRA
SputhAfrica
CHL
COL
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80
MktIncomeGini
ChangeinGinipoints:Post-fiscalvs.Market(declineinGinipointsshowninposi vequadrant)
Results
Redistribution and Poverty Reduction
23
24
-14%
-12%
-10%
-8%
-6%
-4%
-2%
0%
2%
Col
ombia
(201
0)
Per
u(200
9)
Guat
emal
a(20
10)
Cos
ta R
ica(
2010
)
Chile
(200
9)
El S
alva
dor (20
11)
Mex
ico(
2010
)
Bol
ivia
(200
9)
Uru
guay
(200
9)
Bra
zil(20
09)
South
Afr
ica(
2010
)
Change in Headcount Ratio ($2.5 PPP/Day): Disposable vs. Market Income
(in percentage points)
Direct Transfers (net of direct taxes) reduce poverty (except in Ethiopia): CEQ 16
Indirect Taxes increase poverty over and above market income poverty in six out
of the CEQ 16 countries
25
26
0
2000
4000
6000
8000
10000
12000
14000
16000
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
SA
(2
01
0)
Co
l(2
01
0)
Pe
ru
(2
00
9)
Br
a(
20
09
)
Me
x(2
01
0)
Ch
i(2
00
9)
ChangesinHeadcountRatio:Post- iscalvs.MarketIncome(PovertyLine:US$2.50ppp/day)
HeadcountofMarketIncome HeadcountofDisposableIncome HeadcountofPost- iscal GNIpercapita(2005PPP)
South Africa
Brazil
27
-0.09
-0.08
-0.07
-0.06
-0.05
-0.04
-0.03
-0.02
-0.01
0.00
0.01Guatemala(2010)
Bolivia(2009)
Indonesia(2012)
SriLanka(2009)
Colombia(2010)
ElSalvador(2011)
Peru(2009)
Ethiopia(2011)
CostaRica(2010)
Jordan(2010)
Mexico(2010)
Armenia(2011)
Uruguay(2009)
Brazil(2009)
Chile(2009)
SouthAfrica(2010)
ChangeinGini:MarginalContribu onofNetIndirectTaxes
(inGINIpoints)
Contribu onfromNetIndirectTax GiniChange:Post-fiscalvsMkt
Note that Net Indirect Taxes can be equalizing and yet poverty increasing: Ethiopia
0.026
-0.015
0.045
-0.007
0.006 0.006
-0.012 -0.016
-0.002
-0.024-0.019
0.006
0.036
-0.028
0.012
-0.072-8%
-6%
-4%
-2%
0%
2%
4%
6%
8%Ethiopia(2011)
Indonesia(2012)
SriLanka(2009)
Peru(2009)
Colombia(2010)
Guatemala(2010)
CostaRica(2010)
Chile(2009)
ElSalvador(2011)
Jordan(2010)
Mexico(2010)
Bolivia(2009)
Armenia(2011)
Uruguay(2009)
Brazil(2009)
SouthAfrica(2010)
ChangeinHeadcountRa o($2.5PPP/Day):MarginalContribu onfromNetIndirectTaxes(inpercentagepoints)
Contribu onfromNetIndirectTaxes ChangeinHeadcount:Post-FiscalvsMkt
Thank you!
28
